Show simple item record

dc.contributor.authorAcer, Seheren_US
dc.contributor.authorSelvitopi, Oğuzen_US
dc.contributor.authorAykanat, Cevdeten_US
dc.coverage.spatialSantiago de Compostela, Spain
dc.date.accessioned2018-04-12T11:46:45Z
dc.date.available2018-04-12T11:46:45Z
dc.date.issued2017-08-09en_US
dc.identifier.urihttp://hdl.handle.net/11693/37650
dc.descriptionDate of Conference: 28 August - 1 September, 2017
dc.descriptionConference name: Euro-Par: European Conference on Parallel Processing - Euro-Par 2017: Parallel Processing 23rd International Conference on Parallel and Distributed Computing
dc.description.abstractThe scalability of sparse matrix-vector multiplication (SpMV) on distributed memory systems depends on multiple factors that involve different communication cost metrics. The irregular sparsity pattern of the coefficient matrix manifests itself as high bandwidth (total and/or maximum volume) and/or high latency (total and/or maximum message count) overhead. In this work, we propose a hypergraph partitioning model which combines two earlier models for one-dimensional partitioning, one addressing total and maximum volume, and the other one addressing total volume and total message count. Our model relies on the recursive bipartitioning paradigm and simultaneously addresses three cost metrics in a single partitioning phase in order to reduce volume and latency overheads. We demonstrate the validity of our model on a large dataset that contains more than 300 matrices. The results indicate that compared to the earlier models, our model significantly improves the scalability of SpMV. © 2017, Springer International Publishing AG.en_US
dc.language.isoEnglishen_US
dc.source.titleEuro-Par 2017: Parallel Processing - 23rd International Conference on Parallel and Distributed Computingen_US
dc.relation.isversionofhttps://doi.org/10.1007/978-3-319-64203-1_45en_US
dc.subjectCommunication costen_US
dc.subjectHypergraph partitioningen_US
dc.subjectOne-dimensional partitioningen_US
dc.subjectSparse matrix-vector multiplicationen_US
dc.titleAddressing volume and latency overheads in 1d-parallel sparse matrix-vector multiplicationen_US
dc.typeConference Paperen_US
dc.departmentDepartment of Computer Engineeringen_US
dc.citation.spage625en_US
dc.citation.epage637en_US
dc.identifier.doi10.1007/978-3-319-64203-1_45en_US
dc.publisherSpringeren_US
dc.contributor.bilkentauthorAykanat, Cevdet


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record